Post-stroke, functional visual problems are frequently cerebral rather than ocular. However, brain systems are not the basis of current rehabilitation. Even marginally better outcomes in estimated 230,000 US acute stroke survivors with spatial neglect could result in annual savings >$200 million. The Candidate proposed that better spatial neglect rehabilitation outcomes may result from appropriate subject stratification: performance supported by dopaminergic "aiming" systems, as well as established perceptual "where" brain networks, determined recovery in classical animal studies. With outstanding mentoring and training, under a K08 award she developed and validated a spatial neglect assessment method quantifying these recovery components. Novel K02 studies further demonstrated that "aiming" and "where" functional stratification correlated with recovery and response to clinical treatment. Continuous NIH funding since 1999 allowed her to apply the translational stratification hypothesis to neurorehabilitation research, attracting numerous clinician-researcher trainees for hands-on training. Recently, R01 funding allows her to expand her research goals, investigating translational stratification of recovery trajectory in larger groups of stroke survivors with spatial neglect. A physician cognitive neuroscientist applying psychological theory to stroke rehabilitation, she is a unique role model for medical trainees at three distinct levels. Medical students compete for research laboratory rotations, and are mentored individually in clinical research. Physiatric resident physicians are mentored in clinical research, and physiatric residents and post-doctoral fellows perform program-required research in her laboratory, resulting in trainee-authored manuscripts and presentations (5 in 2008). The K24 mechanism provides critical protected time for the Candidate to expand the theoretical basis of her clinical research, and commit appropriately to mentoring activities. The proposal activities also build her collaboration with Anne Foundas, MD. Their developing translational neuro- anatomic model, to potentially predict recovery and response to spatial neglect treatment after stroke, is expected to generate novel hypotheses for further trainee-mentor collaboration and grant applications.